Phase III of the Drosophila genome project will lead to the development of new sequencing technology as well as complete elucidation of the genomic sequence of the 300 kb region that corresponds to the Drosophila bithorax complex (BX-C). The new sequencing methodology is likely to be faster and less expensive than current techniques. In addition, these approaches promise to minimize existing problems in data assembly. Furthermore, it should be possible to perform some of the rate-limiting steps in parallel, making it reasonable to expect that these methodologies could be effectively scaled-up to initiate sequencing the entire Drosophila genome. At the same time, the sequence obtained from the BX-C is likely to play a central role in achieving a better understanding of one of the major molecular mechanisms underlying pattern formation. Recent innovations in two key areas have made it possible to begin the consideration of large-scale sequence analysis. Specifically, new procedures have been developed to generate nearly complete physical maps of entire genomes. The overlapping set of clones upon which these maps are based can be used as substrates in large-scale sequencing efforts. The second area of technology improvement is the development of automated fluorescent sequenators. The major remaining rate-limiting step is the ability to break down the inserts of the mapping clones so that they can be delivered to the sequencing machines in an organized and non-redundant fashion. The minimization of this limitation is a major focus of these phase HI experiments. Two new strategies have been developed that can be used in a time-and cost-effective fashion to sequence intermediate-sized fragments of 5-10 kb. These schemes are based on the introduction, within a target, of fixed priming sites every few hundred base pairs. One approach is based on the forced insertion of transposons into the target sequences and the use of the polymerase chain reaction (PCR) to identify appropriate templates prior to sequencing. The second procedure involves the production of an ordered set of deletions within a target from a fixed point. Such subclones can then be mapped and sequenced from the fixed point. In the proposed phase HI experiments, the 100-200 kb inserts of the mapping clones (Pls and YACs) will be fragmented to intermediate size (5-10 kb) and subcloned. This set of clones will then be ordered to minimize redundancy and sequenced by using the transposon-based and ordered deletion strategies in conjunction with automated fluorescent sequencing.